AMD Radeon 680M vs NVIDIA GeForce RTX 2060 SUPER Comparison

Where Each One Wins

The benchmark database splits these two GPUs cleanly by workload class, and the results are lopsided. The NVIDIA GeForce RTX 2060 SUPER wins every recorded head-to-head test, taking all three comparisons with commanding margins. The AMD Radeon 680M, an integrated graphics processor, does not record a single win in the shared benchmark suite.

The RTX 2060 SUPER's strongest showing comes in the 3DMark Steel Nomad DX12 test, where it scores 2011 against the 680M's 378. That is a 432% advantage, the largest delta in the dataset. In compute-oriented workloads, the gap narrows but remains decisive: the GeForce leads by 227.9% in Geekbench OpenCL and 252.4% in Geekbench Vulkan.

For use-case segmentation, the data points to a clear divide. The RTX 2060 SUPER is a discrete, dual-slot graphics card designed for sustained gaming and content creation on a desktop. Its average benchmark score of 18093 places it in the 62nd percentile of all GPUs. The Radeon 680M, with an average score of 15270, sits in the 57th percentile, which is respectable for an integrated part but a full tier below the NVIDIA card in raw output.

The 680M's role is efficiency and portability rather than peak performance. It draws 50 W, uses no power connectors, and is an IGP (integrated graphics processor) with a PCIe 4.0 x8 interface. The RTX 2060 SUPER, by contrast, requires a 175 W TDP, a single 8-pin connector, and a suggested 450 W PSU. The practical takeaway: the 680M wins on integration and power envelope, while the 2060 SUPER wins on every measured performance metric.

Architecture Differences

The two chips come from different design philosophies and process generations. The RTX 2060 SUPER uses the TU106 chip built on Turing architecture, fabricated by TSMC on a 12 nm process. It packs 10,800 million transistors across a 445 mm² die, yielding a transistor density of 24.3M per mm². The Radeon 680M uses the Rembrandt+ chip with RDNA 2.0 architecture, also from TSMC but on a 6 nm node. It integrates 13,100 million transistors on a much smaller 208 mm² die, achieving 63.0M transistors per mm².

That density difference reflects the 680M's integrated nature. It shares a die with the CPU and memory controllers, so the GPU portion is compact. The 2060 SUPER is a standalone graphics processor with its own dedicated memory subsystem.

The compute resources differ substantially. The RTX 2060 SUPER has 2176 shading units, 136 texture mapping units, 64 render output units, 34 ray tracing cores, and 272 tensor cores. The Radeon 680M has 768 shading units, 48 TMUs, 32 ROPs, and 12 ray tracing cores, but no tensor cores. In raw throughput, the NVIDIA card delivers 7.181 TFLOPS FP32 and 14.36 TFLOPS FP16 (2:1), while the AMD part offers 3.379 TFLOPS FP32 and 6.758 TFLOPS FP16 (2:1).

Memory architecture is another fundamental split. The 2060 SUPER uses 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s bandwidth. The 680M has no dedicated memory; it uses system shared memory with a system-dependent bandwidth. The clock speeds tell a similar story: the NVIDIA card runs at 1470 MHz base and 1650 MHz boost, while the AMD IGP runs at 2000 MHz base and 2200 MHz boost. Higher clocks on the 680M cannot compensate for the massive difference in shading units and memory bandwidth.

Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity is not a differentiator. The RTX 2060 SUPER also carries Turing-specific features like tensor cores for AI workloads, which the Radeon lacks entirely.

Head-to-Head Benchmarks

The recorded head-to-head data contains three tests, and the RTX 2060 SUPER wins all of them. The largest margin appears in 3DMark Steel Nomad DX12: 2011 for the GeForce versus 378 for the Radeon, a 432% difference. That test stresses modern DX12 rendering with ray tracing effects, an area where the 2060 SUPER's dedicated RT cores and tensor hardware give it a structural advantage.

Geekbench OpenCL shows the NVIDIA card scoring 76957 against 23468 for the AMD part, a 227.9% lead. OpenCL compute favors the wider processor: the 2060 SUPER has nearly three times the shading units and more than double the FP32 throughput. The 680M's 2200 MHz boost clock helps it close some ground, but the raw execution width gap is too large.

Geekbench Vulkan produces a similar result: 77402 for the GeForce versus 21965 for the Radeon, a 252.4% advantage. Vulkan tends to scale well with shader count, and the 2176 shading units of the 2060 SUPER overwhelm the 768 of the 680M.

Contextualizing against the nearest rivals clarifies where each part sits. The RTX 2060 SUPER's average score of 18093 is just 0.4% behind the NVIDIA GeForce RTX 3060 Mobile (18159) and 0.5% behind the AMD Radeon Pro 5700 (18189). It is 1.5% behind the AMD Radeon RX 460 (18373) and 1.6% behind the Intel Arc A770M (18383). Those are tight margins, meaning the 2060 SUPER is competitive with a range of midrange discrete GPUs.

The Radeon 680M's average score of 15270 sits 0.1% behind the NVIDIA GeForce GTX 580 (15283) and 0.1% behind the NVIDIA GeForce RTX 2060 (15290). It is 0.5% ahead of the RTX 3050 OEM (15199) and 0.7% ahead of the AMD Radeon RX 7600 (15171). This places the integrated part in the same performance band as older discrete cards, which is notable for an IGP but still far from the 2060 SUPER's league.

Specification Differences

The specification sheet shows where the two diverge most sharply.

• Process node: RTX 2060 SUPER is 12 nm; Radeon 680M is 6 nm.
• Transistors: 10,800 million versus 13,100 million.
• Die size: 445 mm² versus 208 mm².
• Transistor density: 24.3M / mm² versus 63.0M / mm².
• Base clock: 1470 MHz versus 2000 MHz.
• Boost clock: 1650 MHz versus 2200 MHz.
• Memory: 8 GB GDDR6 with 256-bit bus and 448.0 GB/s bandwidth versus system shared memory with system-dependent bandwidth.
• Shading units: 2176 versus 768.
• TMUs: 136 versus 48.
• ROPs: 64 versus 32.
• RT cores: 34 versus 12.
• Tensor cores: 272 versus none.
• Pixel rate: 105.6 GPixel/s versus 70.40 GPixel/s.
• Texture rate: 224.4 GTexel/s versus 105.6 GTexel/s.
• FP32: 7.181 TFLOPS versus 3.379 TFLOPS.
• FP16: 14.36 TFLOPS (2:1) versus 6.758 TFLOPS (2:1).
• TDP: 175 W versus 50 W.
• Slot width: Dual-slot versus IGP.
• Power connectors: 1x 8-pin versus none.
• Suggested PSU: 450 W versus not applicable.
• Bus interface: PCIe 3.0 x16 versus PCIe 4.0 x8.
• Display outputs: 1x DVI, 1x HDMI 2.0, 2x DisplayPort 1.4a, 1x USB Type-C versus portable device dependent.
• Dimensions: 229 mm length, 113 mm height, 35 mm width versus not specified.
• Production status: End-of-life versus active.
• Release date: 2019-07-08 versus 2023-01-02.
• Predecessor: GeForce 10 versus Vega II IGP.
• Successor: GeForce 30 versus Navi III IGP.

The RTX 2060 SUPER has a launch MSRP of 399 USD. The Radeon 680M has no listed launch MSRP.

FAQ

Q: Which GPU wins in raw performance?

A: The NVIDIA GeForce RTX 2060 SUPER wins all three recorded head-to-head benchmarks, with margins ranging from 227.9% in Geekbench OpenCL to 432% in 3DMark Steel Nomad DX12.

Q: How does the Radeon 680M compare to older discrete GPUs?

A: The 680M's average benchmark score of 15270 is 0.1% behind the GeForce GTX 580 and 0.1% behind the GeForce RTX 2060, placing it in the same performance band as those older discrete cards.

Q: What is the main architectural difference?

A: The RTX 2060 SUPER uses Turing architecture on a 12 nm node with 2176 shading units and 272 tensor cores. The Radeon 680M uses RDNA 2.0 on a 6 nm node with 768 shading units and no tensor cores.

Q: Does the Radeon 680M have dedicated memory?

A: No. It uses system shared memory with system-dependent bandwidth, while the RTX 2060 SUPER has 8 GB of GDDR6 on a 256-bit bus with 448.0 GB/s bandwidth.

Q: Which GPU supports ray tracing?

A: Both support DirectX 12 Ultimate (12_2) and include ray tracing cores, but the RTX 2060 SUPER has 34 RT cores versus 12 on the Radeon 680M.

Q: What is the power requirement difference?

A: The RTX 2060 SUPER has a 175 W TDP and requires a 450 W suggested PSU with one 8-pin connector. The Radeon 680M has a 50 W TDP, uses no power connectors, and requires no suggested PSU.